NielsRogge

DINOv2 is now available in HF Transformers (with tutorial)

18

Hi folks, As there are multiple issues here regarding fine-tuning DINOv2 on custom data, questions related to semantic segmentation/depth estimation, image similarity and feature extraction etc. this should now become...

Add integration with Hugging Face

2

Hi, Thanks for this nice work. I wrote a quick PoC to showcase that you can easily have integration so that you can automatically load the various VisionMamba models using...

Add Hugging Face integration

3

Hi, Thanks for this nice work. I wrote a quick PoC to showcase that you can easily have integration so that you can automatically load the various Depth Anything models...

Tutorial: fine-tune SAM on custom dataset (demo notebooks)

7

Hi, SAM was recently added to Hugging Face Transformers 🤗 , and I made some tutorial notebooks to showcase how to fine-tune the model on a different domain (medical images)...

Add LLaVa 1.6

1

# What does this PR do? This PR adds the new LLaVa 1.6 model. To do: - [x] not sure how batched generation works - [x] make `image_sizes` a tensor...

[Docs] Add resources

1

# What does this PR do? This PR adds some more resources regarding various models.

[Feature request] Add direction parameter for truncation

1

Currently, there's a `padding_side` option implemented in HuggingFace Transformers, which lets users pad input from the left or the right. To create fast tokenizers (backed by HuggingFace Tokenizers), this is...

Hey, thanks for this great model! I though it'd be nice to leverage the [PyTorchModelHubMixin](https://huggingface.co/docs/hub/models-uploading#upload-a-pytorch-model-using-huggingfacehub) class, which allows to add `from_pretrained` and `push_to_hub` capabilities to any custom `nn.Module`, which means...

Add save_state_dict

1

This PR adds `save_torch_state_dict` as requested in #2065. To do: - [ ] add `load_torch_state_dict`

This PR improves the documentation of PyTorchModelHubMixin, and also discusses some potential improvements. To do: - [x] add safetensors serialization by default - [x] add `max_shard_size` support - [ ]...